agentleFS
Sign inStart free
ForgeyClapForgeyClap/Forge-Core/.claude/skills/forge-voice/SKILL.md

forge-voice

Skill2 starsChanged 8 days agoMITView on GitHub

Playbook for AI voice and phone agents: Twilio, Vapi (gebruik bij een belagent).

1 other project uses a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use for: AI voice and phone agents (Dutch client phone agents: booking, reception, callback).

    line 11
  2. Very specific:

    Not for: the knowledge/answer layer (forge-rag) or post-call e-mail/CRM (forge-integration).

    line 12
  3. Very specific:

    No outbound/SMS/customer traffic enabled without owner approval (verify none was auto-enabled).

    line 42

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Forge playbook — AI voice / phone agent

When to use

  • Use for: AI voice and phone agents (Dutch client phone agents: booking, reception, callback).
  • Not for: the knowledge/answer layer (forge-rag) or post-call e-mail/CRM (forge-integration).
  • Domain framing, team roles and honesty rule: references/overview.md.

Hard rules

  • Dutch-first when configured for a Dutch client.
  • One question at a time.
  • Natural turn-taking + barge-in.
  • AI disclosure + recording consent (EU/NL).
  • One email per call maximum; idempotency before any notification.
  • No outbound calls and no SMS without explicit owner approval.
  • Honest fallback — no hallucinated business facts.

Full text of every rule: references/hard-rules.md.

Team (conditional)

Lead: integration-boss (telephony wiring + notifications) with head-chef/build-boss if there's an app around it. Full roster: references/team-and-tools.md.

Skills / commands / MCP

forge-rag (grounded answers + citations for the knowledge layer), forge-integration (post-call e-mail/CRM, draft-gated + idempotent), humanizer (natural Dutch phrasing, on-demand), telephony/voice SDK docs via Context7 (Twilio / Vapi / Retell / LiveKit / ElevenLabs — exact turn-taking + barge-in config). claude-api if the dialogue LLM is Claude.

Fan-out & flow

L2 single inbound flow (answer → intent → one action); L3 multi-intent agent + booking + notifications + knowledge layer. Parallel: telephony/turn-taking wiring ∥ knowledge/answer layer (forge-rag) ∥ post-call notification (forge-integration) — independent once the conversation-state contract is fixed. Serial: conversation design (states, one-question-per-turn, Dutch prompts) → AI disclosure + consent → intent/slot filling → action (book/callback) → single idempotent notification → clean close.

Domain gates

  • Dutch-first caller experience verified (spoken-natural, correct voice) when configured for a Dutch client.
  • Exactly one question per turn; barge-in works and cuts TTS within ~200ms; latency measured, not assumed.
  • AI disclosure present + early; recording consent basis recorded; opt-out honoured immediately.
  • ≤1 email per call; notification path is idempotent (a retried turn sends zero extra mails); lead capture consent-aware.
  • No outbound/SMS/customer traffic enabled without owner approval (verify none was auto-enabled).
  • Unknown → honest fallback + human handoff; no invented business facts; secrets in env.

Ship-readiness (unique)

Evidence checklist before handoff: references/ship-readiness.md.

Safety, approval and honesty

  • Approval / safety: no outbound calls and no SMS without explicit owner approval; at most one email per call, idempotent.
  • Approval / safety: AI disclosure and recording consent (EU/NL).
  • Honest failure: never claim a live call worked unless a real call was placed and observed; label anything not exercised not-run.
  • Untrusted content: see references/untrusted-content.md.
  • Shared gates for every playbook (honesty, owner approval, secrets, reviews): .claude/docs/forge-reference/playbook-template.md.

References

  • references/overview.md — domain framing paragraph(s) from under the title.
  • references/hard-rules.md — full text of every hard rule.
  • references/team-and-tools.md — full team roster, skills, commands, MCP and opt-in runtimes.
  • references/ship-readiness.md — full ship-readiness evidence checklist.
  • references/untrusted-content.md — Untrusted-content note.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free